Search

In intrusion detection, the curse of dimensionality and the trade-off between maintaining a low false alarm rate and achieving a high detection rate are significant challenges. This research suggests a unique strategy based on dimensionality reduction methods to improve the performance of network intrusion detection systems (NIDS). Compressing high-dimensional network traffic data using a Long Short-Term Memory Autoencoder (LSTMAE) allows the reduced characteristics to be submitted to a classifier to identify anomalies that may indicate an attack. Using standard datasets, including Network Security Laboratory - Knowledge Discovery in Datasets (NSL-KDD), UNSW-NB15, and Canadian Institute for Cyber Security - Intrusion Detection Systems (CICIDS2017), the proposed model is tested with classifiers like Random Forest (RF) and LightGBM (Light Gradient Boosting Machines). It is hoped that by adopting this method, NIDS response times may be improved while costs associated with storing and processing data are minimized. Precision, recall, F-score, accuracy, detection rate (DR), and false alarm rate (FAR) are only a few of the performance measures used to assess the quality of the suggested models. The experimental findings show that the proposed LSTMAE model reduces prediction errors more effectively than classic machine learning techniques such as Random Forest (RF), Gradient Boosting (GB), Support Vector Machines (SVM), Deep Belief Networks (DBN), Deep Neural Networks (DNN), Autoencoder (AE), and Long Short-Term Memory (LSTM). The results also show that the proposed solution outperforms the state-of-the-art methods of detection accuracy and computing complexity using accuracy, precision, recall, F1_Score, detection rate, and FAR. [ABSTRACT FROM AUTHOR]

Rainfall forecasting is essential because heavy and irregular rainfall creates many impacts like destruction of crops and farms. Here, the occurrence of rainfall is highly related to atmospheric parameters. Thus, a better forecasting model is essential for an early warning that can minimize risks and manage the agricultural farms in a better way. In this manuscript, Deep Neural Network (DNN) optimized with Flamingo Search Optimization Algorithm (FSOA) is proposed for Long-term and Short-term Rainfall forecasting. Here, the rainfall data is obtained from the standard dataset as Sudheerachary India Rainfall Analysis (IRA). Moreover, the Morphological filtering and Extended Empirical wavelet transformation (MFEEWT) approach is utilized for pre-processing process. Also, the deep neural network is utilized for performing rainfall prediction and classification. Additionally, the parameters of the DNN model is optimizing by Flamingo Search Optimization Algorithm. Finally, the proposed MFEEWT-DNN- FSOA approach has effectively predict the rainfall in different locations around India. The proposed model is implemented in Python tool and the performance metrics are calculated. The proposed MFEEWT-DNN- FSOA approach has achieved 25%, 26%, 25.5% high accuracy and 35.8%, 24.7%, 15.9% lower error rate for forecasting rainfall in Cannur at Kerala than the existing Map-Reduce based Exponential Smoothing Technology for rainfall prediction (MR-EST-RP), modular artificial neural networks with support vector regression for rainfall prediction (MANN-SVR-RP), and biogeography-based extreme learning machine (BBO-ELM) (BBO-ELM-RP) methods respectively. [ABSTRACT FROM AUTHOR]

Software defect prediction is used to assist developers in finding potential defects and allocating their testing efforts as the scale of software grows. Traditional software defect prediction methods primarily concentrate on creating static code metrics that are fed into machine learning classifiers to predict defects in the code. To achieve the desired classifier performance, appropriate design decisions are required for deep neural network (DNN) and convolutional neural network (CNN) models. This is especially important when predicting software module fault proneness. When correctly identified, this could help to reduce testing costs by concentrating efforts on the modules that have been identified as fault prone. This paper proposes a CONVSDP and DNNSDP (cognitive and neural network) approach for predicting software defects. Python Programming Language with Keras and TensorFlow was used as the framework. From three NASA system datasets (CM1, KC3, and PC1) selected from PROMISE repository, a comparative analysis with machine learning algorithms (such as Random Forest (RF), Decision Trees (DT), Nave Bayes (NF), and Support Vector Machine (SVM) in terms of F-Measure (known as F1-score), Recall, Precision, Accuracy, Receiver Operating Characteristics (ROC) and Area Under Curve (AUC) has been presented. We extract four dataset attributes from the original datasets and use them to estimate the development effort, development time, and number of errors. The number of operands, operators, branch count, and executable LOCs are among these attributes. Furthermore, a new parameter called cognitive weight (Wc) of Basic Control Structure (BCS) is proposed to make the proposed cognitive technique more effective, and a cognitive data set of 8 features for NASA system datasets (CM1, KC3, and PC1) selected from the PROMISE repository to predict software defects is created. The experimental results showed that the CONVSDP and DNNSDP models was comparable to existing classifiers in both original datasets and cognitive data sets, and that it outperformed them in most of the experiments. [ABSTRACT FROM AUTHOR]

This study is devoted to the development of an Adaptive-Neuro-Fuzzy-Inference-System (ANFIS) model for the prediction of ultimate load of rectangular concrete-filled steel tubular structural members. The learning process of the model is performed by conducting a combination of backpropagation gradient descent and least-squares techniques. The performance of the model is examined by several quality metrics such as coefficient of determination (R2), Root-Mean-Squared-Error (RMSE), Mean-Absolute-Error (MAE), Index of Agreement (IA) and Slope of linear regression. Monte Carlo random sampling technique is employed to propagate input variations to the output response. Moreover, the performance of ANFIS is also compared with other machine learning models including Artificial Neural Network (ANN), Support Vector Machine (SVM), Gaussian Process Regression (GPR) and Ensemble. Results show that the ANFIS model yields higher prediction performance than other machine learning models, for both training and testing data points and regarding all quality metrics. For instance, using training data points, the ANFIS model exhibits a RMSE of 0.0283 compared to 0.0342, 0.0588, 0.0291, and 0.0464 using ANN, Ensemble, GPR, and SVM, respectively (the corresponding gain values are+17.3%,+51.9%,+2.8%, and+39.0%, respectively). On the other hand, using testing data points, the ANFIS model exhibits a RMSE of 0.0276 compared to 0.0393, 0.0987, 0.0403, and 0.0460 using ANN, Ensemble, GPR, and SVM, respectively (the corresponding values of gain are+29.8%,+72.1%,+31.5%, and+40.0%, respectively). The same observation can be made for other quality metrics. It can be concluded that the ANFIS model outperforms other models for both training and testing datasets. The ANFIS model is also compared with existing works in the past, showing its improvement in prediction results. Finally, sensitivity analysis is performed to determine the degree of effect of the input parameters on the ultimate load. [ABSTRACT FROM AUTHOR]

Focal cortical dysplasia (FCD) is an inborn anomaly in brain growth and morphological deformation in lesions of the brain which induces focal seizures. Neurosurgical therapies were performed for the detection of FCD. Furthermore, it can be overcome through the presurgical evaluation of epilepsy. The surgical result is attained basically through the output of the presurgical output. In preprocessing the process of increasing true positives with the decrease in false negatives occurs which results in an effective outcome. MRI (Magnetic Resonance Imaging) outputs are efficient to predict the FCD lesions through T1- MPRAGE and T2- FLAIR efficient output can be obtained. In our proposed work we extract the S2 features through the testing of T1, T2 images. Using RNN-LSTM (Recurrent neural network-Long short-term memory) test images were trained and the FCD lesions were segmented. The output of our work is compared with the proposed work yields better results compared to the existing system such as artificial neural network (ANN), support vector machine (SVM), and convolution neural network (CNN). This approach obtained an accuracy rate of 0.195% (ANN), 0.20% (SVM), 0.14% (CNN), specificity rate of 0.23% (ANN), 0.15% (SVM), 0.13% (CNN) and sensitivity rate of 0.22% (ANN), 0.14% (SVM), 0.08% (CNN) respectively in comparison with RNN-LSTM. [ABSTRACT FROM AUTHOR]

This paper refers to a new resilient cyber-physical machine learning-based system that enables the generation of high-resolution tomographic images. The research object was a model of a tank filled with tap water. Using electrical impedance tomography (EIT) with 16 electrodes, the possibility of identifying inclusions inside the reservoir was investigated. A two-stage hybrid approach was proposed. In the first stage, three independent models were trained for the Elastic Net, Artificial Neural Networks (ANN) and Support Vector Machine (SVM) methods. In the second stage, a k-Nearest Neighbors (kNN) classification model was trained, that optimizes tomographic reconstructions by selecting the best method for each pixel, taking into account the specificity of a given measurement vector. Research has shown that applying the new concept results in a higher reconstruction quality than other methods used singly. It should be emphasized that our research is not intended to develop a new homogenous machine learning method. Instead, the goal is to invent an innovative, original, and flexible way to simultaneously use multiple machine learning methods for image optimization in industrial electrical impedance tomography. [ABSTRACT FROM AUTHOR]

Affecting millions of people all over the world, the COVID-19 pandemic has caused the death of hundreds of thousands of people since its beginning. Examinations also found that even if the COVID-19 patients initially survived the coronavirus, pneumonia left behind by the virus may still cause severe diseases resulting in organ failure and therefore death in the future. The aim of this study is to classify COVID-19, normal and viral pneumonia using the chest X-ray images with machine learning methods. A total of 3486 chest X-ray images from three classes were first classified by three single machine learning models including the support vector machine (SVM), logistics regression (LR), artificial neural network (ANN) models, and then by a stacking model that was created by combining these 3 single models. Several performance evaluation indices including recall, precision, F-1 score, and accuracy were computed to evaluate and compare classification performance of 3 single four models and the final stacking model used in the study. As a result of the evaluations, the models namely, SVM, ANN, LR, and stacking, achieved 90.2%, 96.2%, 96.7%, and 96.9%classification accuracy, respectively. The study results indicate that the proposed stacking model is a fast and inexpensive method for assisting COVID-19 diagnosis, which can have potential to assist physicians and nurses to better and more efficiently diagnose COVID-19 infection cases in the busy clinical environment. [ABSTRACT FROM AUTHOR]

Development of a robust triple multimodal biometric approach for human authentication using fingerprint, iris and voice biometric is the main objective of this manuscript. Accordingly, three essential algorithms for biometric authentication are presented. The extracted features from these multimodals are combined via feature fusion center (FFC) and feature scores. These features are trained through artificial neural network (ANN) and support vector machine (SVM) classifiers. The first algorithm depends on boundary energy method (BEM) extracted features from fingerprint, normalized combinational features from iris and dimensionality reduction methods (DRM) from voice using sum/average FFC. The second proposed algorithm uses extracted features from zoning method of fingerprint, SIFT of iris and higher order statistics (HOS) of voice signals. The third proposed algorithm consists of extracted features from zoning method for fingerprint, SIFT from iris and DRM from voice signals. Classification accuracy of implemented algorithms is estimated. Comparison between proposed algorithms is introduced in terms of equal error rate (EER) and ROC curves. The experimental results confirm superiority of second proposed algorithm which achieves a classification rate of 100% using SVM classifier and sum FFC. From computational point of view, the first algorithm consumes the lowest time using SVM classifier. On other hand, the lowest EER is achieved by first proposed algorithm for extracted features from Karhunen-Loeve transform (KLT) method of DRM. Additionally, the lowest ROC curves are accomplished respectively for extracted features from multidimensional scaling (MDS), generated ARMA synthesis and Isomap features. Their accuracy is improved with SVM. Also, the sum FFC introduces efficient results compared to average FFC. These algorithms have the advantages of robustness and the strength of selecting unimodal, double and triple biometric authentication. The obtained results accomplish a remarkable accuracy for authentication and security within multi practical applications. [ABSTRACT FROM AUTHOR]

Cancer is a prevalent disease which comes in several forms. The need of the hour in cancer research is to be able to diagnose cancer in its early stages. The furthermost common forms of cancer among women us breast cancer. In recent times, there has been a drastic increase in the number of breast cancer cases among women. As a wide range of medical data is available in electronic form and with easy access to Machine Learning(ML) techniques disease progression risk evaluation has been made easier. These ML tools can aid in giving us complex insights from the massive amounts of available data. Some of the techniques used for developing predictive models for perfect decision making in cancer research are Artificial Neural Networks (ANNs), Bayesian Networks (BNs), Support Vector Machines (SVMs), and Decision Trees (DTs). Although it is acceptable that ML is used to predict cancer progression, we need some level of validation. In this paper, we have come up with a review of several ML methods in modelling cancer progression. We discuss several predictive models based on supervised ML techniques and the inputs given by users, along with the data available. The results that were obtained from Logistic Regression show us that this method gave a significantly higher accuracy than most other classifiers. The best accuracy is 98.2%, however, the best precision and recall is 100 and 98.60% correspondingly. [ABSTRACT FROM AUTHOR]

Customer behavior prediction is gaining more importance in the banking sector like in any other sector recently. This study aims to propose a model to predict whether credit card users will pay their debts or not. Using the proposed model, potential unpaid risks can be predicted and necessary actions can be taken in time. For the prediction of customers' payment status of next months, we use Artificial Neural Network (ANN), Support Vector Machine (SVM), Classification and Regression Tree (CART) and C4.5, which are widely used artificial intelligence and decision tree algorithms. Our dataset includes 10713 customer's records obtained from a well-known bank in Taiwan. These records consist of customer information such as the amount of credit, gender, education level, marital status, age, past payment records, invoice amount and amount of credit card payments. We apply cross validation and hold-out methods to divide our dataset into two parts as training and test sets. Then we evaluate the algorithms with the proposed performance metrics. We also optimize the parameters of the algorithms to improve the performance of prediction. The results show that the model built with the CART algorithm, one of the decision tree algorithm, provides high accuracy (about 86%) to predict the customers' payment status for next month. When the algorithm parameters are optimized, classification accuracy and performance are increased. [ABSTRACT FROM AUTHOR]

One of the momentous transformation performed by an artificial neural network (ANN), Support Vector Machine (SVM), Radial basis Function (RBF) and many other machine learning method is the application of activation function. MyAct the proposed activation method is used here with various ANN architectures for link prediction, classification and general prediction. Statistical properties of data used here to prove the effectiveness of proposed activation function MyAct over other popular activation methods. A data dependent transfer method is developed, which is pioneer in its own way. This proves to be an unified formulation for the robust and generalised learning for the classification, link prediction and regression problem types. Classification is done with Iris dataset using ANN with different activation method and results are compared. Improved results are achieved when MyAct used with Tailored Deep Feed Forward Artificial Neural Network (TDFFANN), simple Artificial Neural Network and Deep Artificial Neural Network. Aim here is to develop a novel activation method which work with positive data, negative data, small size data, big size data, skewed data or corrupt data. An attempt is made to cover complete versatile behaviour of data. Currently not a single activation method can work well on all above mentioned data. Results obtained using MyAct on the datasets used here proves it to be a good choice in comparison to logsig, tansig and other popular activation methods for classification and link prediction. Satisfactory improvement is achieved by using data length as well as negative range values in the prediction done by proposed method. MyAct had 22% better standard deviation than ReLU (Rectified Linear unit) and 36. 28% better standard deviation than ELU (Exponential linear unit). MyAct has 2. 6% better accuracy in regression error than Swiss method and 2. 5% better accuracy in regression error than ELU. Other results are discussed in the paper. [ABSTRACT FROM AUTHOR]

Energy uncertainty and ecological pressures have contributed to a high volatility in energy demand and consumption. The building sector accounts for 30 to 40% of the total global energy consumption. There is a high demand for novel techniques and viable energy strategies for reducing energy consumption in this domain. Energy prediction models have the potential to play a pivotal role in optimising energy consumption. The proposed work presents a new and accurate Energy Demand Prediction (EDP) model for large buildings. This approach leverages the Random Neural Network (RNN) prediction methodology. The proposed RNN-based EDP is compared with traditional Artificial Neural Network (ANN), Support Vector Machine (SVM) and linear regression models. A large building is modelled and simulated for one year in the Integrated Environment Solutions Virtual Environment (IES-VE). Several data inputs such as air temperature, internal gain and the number of people (occupancy) are calculated from IES-VE model and provided to traditional ANN and the proposed RNN predictor. A number of test parameters such as Root Mean Square (RMSE), Normalized Root Mean Square (N-RMSE), Mean Absolute Percentage Error (MAPE) and R provide the proposed RNN model with higher accuracy over the traditional ANN, SVM and linear regression. The proposed RNN predictor provides approximately half of the error of the ANN model. The traditional ANN model gives higher error values of 2.07×, 1.83× and 2.35× for RMSE, NRMSE and MAPE, respectively as compared to the proposed RNN model. Furthermore, the error values of SVM and linear regression were also higher than the proposed EDP scheme. [ABSTRACT FROM AUTHOR]

Small-target categorization of butterflies suffers from large-scale search space of candidate target locations, subtler discriminations, camouflaged appearances, and complex backgrounds. Precise localization and domain-specific discrimination extraction are crucial for this issue. In this work, a novel hierarchical coarse-to-fine convolutional neural network (C-t-FCNN) was proposed. It consists of CoarseNet and FineNet, which incorporate object-level and part-level representations into framework. Specifically, the coarse-grained features containing the orientation description are generated by CoarseNet, while the fine-grained discriminations with semantic distinctiveness are captured by FineNet. Next, the correspondences are established to mark the target regions, background regions, and mismatched regions depending on the quantification of scale-invariant feature transform (SIFT) descriptors. Then, the features are subsampled via spatial pyramid pooling (SPP) for size uniformity and integration. Finally, the irrelevant background and mismatched regions are eliminated by the support vector machine (SVM) with a radial basis function (RBF) kernel, leaving only the target-specific patches for finer-scale extraction. Hence the numeracy can be economized from identifying irrelevant areas and can be rescheduled in feature extraction and final decision, which can suppress time complexity simultaneously. A total of 119,016 augmented butterfly images spanning 47 categories are utilized for model training, while 13,734 images are evaluated for effectiveness verification. The C-t-FCNN delivers impressive performance, i.e., it achieves a validation accuracy of 92.08% and a testing accuracy of 91.6%, which outperforms state-of-the-arts. [ABSTRACT FROM AUTHOR]

Background: Applied research on artificial intelligence, mainly in deep learning, is widely performed. If medical images can be evaluated using artificial intelligence, this could substantially improve examination efficiency.Objective: We investigated an evaluation system for medical images with different noise characteristics using a deep convolutional neural network.Methods: Simulated computed tomography images are the targets of the system. We used an AlexNet trained with natural images for the deep convolutional neural network and a support vector machine for classification. Synthetic computed tomography images with circular and rectangular signal bodies at different levels of contrast and added Gaussian noise were used for training and testing.Results: Two transfer learning methods were tested: classification by a re-trained support vector machine using the AlexNet features, and a method that fine-tuned the deep convolutional neural network. Using the first method, all the test image noise levels could be classified correctly. The fine-tuning method achieved an accuracy rate of 92.6%.Conclusions: An image quality evaluation method using artificial intelligence will be useful for clinical images and different image quality indices in the future. [ABSTRACT FROM AUTHOR]

In order to meet the real-time and robustness requirement of driverless cars driving on highway, this paper proposed a lane line identification method based on the Deep Learning. This method first built a lane line image library and then input the pictures of the lane line image for denoising and normalized processing. Secondly, the Lenet – 5 network model was used for classification and recognition, with a recognition rate of 99.4%, and the lane line type was displayed through GUI interface. Finally, this method was compared with the support vector machine and BP neural network, and the results effectively verified that the method can satisfy the requirement of real-time and accuracy of lane line identification. [ABSTRACT FROM AUTHOR]

We present a method for gender and language variety identification using a convolutional neural network (CNN). We compare the performance of this method with a traditional machine learning algorithm – support vector machines (SVM) trained on character n-grams (n = 3–8) and lexical features (unigrams and bigrams of words), and their combinations. We use a single multi-labeled corpus composed of news articles in different varieties of Spanish developed specifically for these tasks. We present a convolutional neural network trained on word- and sentence-level embeddings architecture that can be successfully applied to gender and language variety identification on a relatively small corpus (less than 10,000 documents). Our experiments show that the deep learning approach outperforms a traditional machine learning approach on both tasks, when named entities are present in the corpus. However, when evaluating the performance of these approaches reducing all named entities to a single symbol "NE" to avoid topic-dependent features, the drop in accuracy is higher for the deep learning approach. [ABSTRACT FROM AUTHOR]

The importance of the surveillance is increasing every day. Surveillance is monitoring of activities, behavior and other changing information. An intelligent automatic system to detect behavior of the human is very important in public places. For this necessity, a framework is proposed to detect suspicious human behavior as well as tracking of human who is doing some unusual activity such as fighting and threatening actions and also distinguishing the human normal activities from the suspicious behavior. The human activity is recognized by extracting the features using the convolution neural network (CNN) on the extracted optical flow slices and pre-training the activities based on the real-time activities. The obtained learned feature creates a score for each input which is used to predict the type of activity and it is classified using multi-class support vector machine (MSVM). This improved design will provide better surveillance system than existing. Such system can be used in public places like shopping mall, railway station or in a closed environment such as ATM where security is the prime concern. The performance of the system is evaluated, by using different standard datasets having different objects and achieved 95% performance as explained in experimental analysis. [ABSTRACT FROM AUTHOR]

In this paper, a hybrid model composed of the correlation ratio analysis and the Support Vector Machine (SVM) is proposed for trip mode recognition. The correlation ratio analysis algorithm is applied to determine the optimal time window of the input attributes so as to optimize the input parameters. The SVM is applied to carry out the trip mode recognition for the whole trip. On this basis, the influence of data sampling frequency on trip mode recognition is further evaluated using large-scale field test data. The results show that: (1) the correlation ratio analysis and the SVM hybrid model attains the best performance for trip mode recognition, the average mode recognition precision reaches 89.8% at the sampling frequency of 1 s; (2) the data sampling frequency significantly affects the trip mode recognition effect; when the sampling frequency is high (less than 5 s), the mode recognition precision is above 70%, however, when the sampling frequency is relatively low (more than 30 s), the bus and car mode recognition precision falls rapidly below 37%. These results provide a reference for using the smartphone sensor datasets to supplement or even replace household travel surveys in transportation planning in the future. [ABSTRACT FROM AUTHOR]

Hyperspectral Image (HSI) store the reflectance values of a single scene or object in several continuous bands of electromagnetic spectrum. When the image is recorded, the information in some of the spectral bands gets mixed with noise. The classification accuracy of hyperspectral image varies inversely with the quantity and nature of noise present in the cluster of spectral bands. Thus, denoising is a fundamental prerequisite in image processing applications like classification, unmixing, etc. In this paper, we compare the effect of denoising via classification using Vectorized Convolutional Neural Network (VCNN), kernel based Support Vector Machine (SVM) and Grand Unified Regularized Least Squares (GURLS) classifiers. The classifiers are provided with raw data (without denoising) and denoised data using spectral and spatial Least Square (LS) techniques. The data given to the network are in the form of pixels, so we call the convolutional neural network (CNN) as VCNN. The experiments are performed on three standard HSI datasets. The performance of the classifiers are evaluated based on overall and class-wise accuracy. [ABSTRACT FROM AUTHOR]

Elephant play a major role in maintaining the ecosystem. Elephants move out of their corridor in search of food and water, result in rise of human-elephant conflicts. Human-Elephant conflict arises in different form such as destruction of field by elephants, elephants are runover by train, elephants getting electrocuted etc. Prevention of elephants entering into human living areas, will reduce the human-elephant conflict to a greater extent. In this paper, performance of conventional image processing techniques, custom Convolutional Neural Network (CNN) architecture and transfer learning based CNN architectures is investigated in the context of human-elephant conflict management system. On identification of elephant from the acquired video feed, the sounds (humming of Bee, Tiger growls) are generated to mitigate the progress of elephant into human living areas. It is observed that the convolutional neural networks produced a higher accuracy or prediction rate compared to conventional image processing technique. It is also observed that, VGG16 CNN model produced an accuracy of 94% with an average computation time varying between 1.5 to 2.1 s. For real time implementation, SqueezeNet CNN model is used because of its lower computation time (varying between 0.02 to 0.05 s) and moderate accuracy of 92.67%. [ABSTRACT FROM AUTHOR]

BACKGROUND: Low-quality medical images may influence the accuracy of the machine learning process. OBJECTIVE: This study was undertaken to compare accuracy of medical image classification among machine learning methods, as classification is a basic aspect of clinical image inspection. METHODS: Three types of machine learning methods were used, which include Support Vector Machine (SVM), Artificial Neural Network (ANN), and Convolution Neural Network (CNN). To investigate changes in accuracy related to image quality, we constructed a single dataset using two different file formats of DICOM (Digital Imaging and Communications in Medicine) and JPEG (Joint Photographic Experts Group). RESULTS: The JPEG format contains less color information and data capacity than the DICOM format. CNN classification was accurate for both datasets, whereas SVM and ANN accuracy decreased with the loss of data from DICOM to JPEG formats. CONCLUSIONS: CNN is more accurate than conventional machine learning methods that utilize the manual feature extraction. [ABSTRACT FROM AUTHOR]

Automatically identifying Chinese characters that are similar in their glyph, pronunciations and meaning are important for building smart question generation tools in a computer-assisted language-learning environment. Previous research on the Chinese character similarity measurement focused on character glyph (e.g. structures, strokes and radicals) with heuristic algorithms whose parameter have preset values. This article presents a machine learning (regression) approach to measure the similarity between two Chinese characters, based on the information which not only includes the glyph, but also pronunciation (pinyin) and semantic meaning derived from HowNet. We evaluated various regression models using a testing set consisting of 2586 pairs of characters selected from elementary Chinese textbooks used. The study results showed that four regression models (M5, Support Vector Machine, Gaussian Process and Linear Regression) have similar results (0.617 ⩽ Mean Absolute Error ⩽ 0.641 , 0.772 ⩽ Root Mean Square Error ⩽ 0.790). In addition, the study implied that the performance of the regression model could be influenced by the character frequency. Moreover, we evaluated the regression model in a well-known Chinese language learning resource, called 100 pairs of the most confusing Chinese characters. The experiment results indicated that this approach has potential in the recognition and generation of confusing Chinese character pairs. [ABSTRACT FROM AUTHOR]

Eye detection plays an important role in many fields, because eyes provide prominent facial feature information. However, changes in face pose, illumination variation, with glasses, and eye occlusions can make it difficult to detect eyes well from facial images. This paper proposes a hybrid model for eye detection. The model is an integration of two classifiers: Convolutional Neural Networks (CNN) and Support Vector Machines (SVM). In order to improve the speed of detection in the system, an eye variance filter (EVF) is constructed for eliminating most of noneye images to keep less candidate eye images. The CNN then works as a trainable feature extractor to explicitly extract various latent eye features. Finally, the trained SVM classifier is employed for eye verification instead of using the CNN classification function. Experiments applying the model have been conducted on the BioID, IMM, FERET and ORL face databases. Comparisons with other methods on the same databases indicate that this hybrid model has achieved a higher detection accuracy. Extensive experiments demonstrate the robustness and efficiency of our method by testing it on different facial images with varying eye conditions. [ABSTRACT FROM AUTHOR]

A novel supervised particle swarm optimization (S-PSO) classification algorithm is proposed for fault diagnosis. In order to improve the accuracy of fault diagnosis and obtain the global optimal solutions with a higher probability, two strategies, i.e. a hybrid particle position updating strategy and a fixed iteration interval intervention updating strategy, are designed to balance the effect of the local and the global search. These methods increase the diversity of particles, expand the particles ability of searching the entire solution space, and guide the particles adaptively jumping out of the local optimal area. Meanwhile, based on the shorter intra-class distance, longer inter-class distance and maximum classification accuracy of training samples, a fitness function is designed to constraint the output optimal class centers. Experimental results demonstrate that the proposed S-PSO classification algorithm can overcome the problems in the classical clustering algorithms, which only consider the similarity of data instead of their physical meanings. The comparison on GE90 engine borescope image texture feature classification is also conducted. The results show that the performance of S-PSO classification algorithm is robust. Its classification accuracy is higher than those of popular methods, including support vector machine (SVM), neural network, Bayesian classifier, and k-nearest neighbor (k-NN) algorithm. [ABSTRACT FROM AUTHOR]

Extreme learning machine (ELM) has demonstrated great potential in machine learning and data mining. Smoothing strategy is an important technology for continuous optimizations. In this work, we apply a smoothing technique to replace the hinge loss function by an accurate smooth approximation. This will allow us to solve ELM as an unconstrained minimization problem directly. We term this reformulated problem as smooth ELM (SELM). A Newton-Armijo algorithm is used to solve the proposed SELM, and the resulting algorithm converges globally and quadratically. The proposed SELM with fast running speed has less decision variables and can better deal with nonlinear problems than the existing smooth support vector machine. Numerical experiments on various types of datasets including two-class datasets and multi-class datasets demonstrate that the speed of SELM is much faster than that of the existing ELM models. And compared with other popular algorithms of support vector machine and ELM, the proposed SELM achieves better or similar generalization. These demonstrate the effectiveness and fast speed of the algorithm. [ABSTRACT FROM AUTHOR]

This paper offers a recurrent neural network to support vector machine (SVM) learning in regression arising widespread applications in a variety of setting. The SVM learning problem in regression is first converted into an equivalent quadratic programming (QP) formulation. An artificial neural network for SVM learning is then proposed. The presented neural network framework guarantees to obtain the optimal solution of the support vector regression (SVR). The existence and convergence of the trajectories of the network are studied. The Lyapunov stability for the considered neural network is also shown. Two illustrative examples provide a further demonstration of the effectiveness of the method. [ABSTRACT FROM AUTHOR]

One-Class Classifier (OCC) has been widely used for its ability to learn without counterexamples. Its extension for multi-class implementation offers an open scheme which allows easily adding new classes. However, using OCCs for the multi-class implementation usually achieves less accuracy than the usual multi-class implementations. Hence, in order to improve the accuracy and keep the offered advantage, the suitable approach consists to combine different classifiers. Thus, this paper proposes a study of combining different types of OCC for multi-class classification by means of a new Dynamic Weighted Average (DWA) combination rule. Experimental results conducted on several real-world datasets prove the effective use of the proposed approach where the DWA rule achieves the best results against fixed rules as well as the decision template. Furthermore, comparison of the proposed open classification system against a standard open classifier based on K-Nearest Neighbor (K-NN) shows the superiority of the proposed system. [ABSTRACT FROM AUTHOR]

Background: Detection of clustered microcalcification (MC) from mammograms plays essential roles in computer-aided diagnosis for early stage breast cancer.Objective: To tackle problems associated with the diversity of data structures of MC lesions and the variability of normal breast tissues, multi-pattern sample space learning is required.Methods: In this paper, a novel grouped fuzzy Support Vector Machine (SVM) algorithm with sample space partition based on Expectation-Maximization (EM) (called G-FSVM) is proposed for clustered MC detection. The diversified pattern of training data is partitioned into several groups based on EM algorithm. Then a series of fuzzy SVM are integrated for classification with each group of samples from the MC lesions and normal breast tissues.Results: From DDSM database, a total of 1,064 suspicious regions are selected from 239 mammography, and the measurement of Accuracy, True Positive Rate (TPR), False Positive Rate (FPR) and EVL = TPR* 1-FPR are 0.82, 0.78, 0.14 and 0.72, respectively.Conclusion: The proposed method incorporates the merits of fuzzy SVM and multi-pattern sample space learning, decomposing the MC detection problem into serial simple two-class classification. Experimental results from synthetic data and DDSM database demonstrate that our integrated classification framework reduces the false positive rate significantly while maintaining the true positive rate. [ABSTRACT FROM AUTHOR]

Within the field of data mining and machine learning, the K-Nearest Neighbor algorithm is a classic algorithm which simply yet elegantly classifies data based upon its similarity to other data. While it follows that the accuracy increases as more data are provided, handling large sets of data is difficult to process serially. It is therefore ideal to perform these tasks in parallel or distributed mode. In this paper, we proposed a framework for distributed nearest neighbor classification. A custom K-Nearest Neighbor algorithm was developed using Hadoop, an environment for developing and deploying applications in parallel on a cluster. The algorithm was implemented on a cluster then tested for accuracy and time of execution. It was observed that the accuracy depends on the provided k-value and on the data set, which is to be expected for the K-Nearest Neighbor process. The time of execution was found to increase logarithmically as the file size, and thus the amount of data the algorithm must parse, increases exponentially. [ABSTRACT FROM AUTHOR]

Since the traditional probabilistic neural network (PNN) cannot systematically solve the difficulty of estimating probability function and the high space complexity, this paper introduces backpropagation (BP) algorithm into the classical PNN. By designing appropriate error function and BP algorithm based on the steepest descent, an improved BP-PNN is presented, with its algorithm and effectiveness deduced. Three synthetic datasets and ten benchmark problems have been tested, compared with Probabilistic Neural Networks (PNN), Multi-Layered Perceptron (MLP) and Support Vector Machine (SVM). The results prove that (1) the accuracy of classification of BP-PNN is much higher than PNN, and it has a significant advantage compared with MLP and SVM; (2) BP-PNN has strong capacity to identify the importance of input indicators; (3) BP-PNN is a new pattern classification method to estimate the probabilistic function, reduce the space complexity and identify the importance of the indicators. [ABSTRACT FROM AUTHOR]

The productivity of steam flooding production wells is determined by numerous factors. In the paper the dominant influence factors of the productivity have been revealed using Gray Relation Analysis (GRA). Based on the Gray Relation Analysis, a model was established for predicting the productivity of steam flooding production wells by Support Vector Machine (SVM), the training algorithm was used to train the model and predict the productivity of test samples. The results show that high precision can be achieved from SVM, the error committed by the model was about ± 10% which is acceptable in field application. The results of SVM were compared with those of Group Method of Data Handling (GMDH) and Backward Propagation (BP) Artificial Neural Network, it indicated that the prediction precision of SVM was obviously higher than that of the two mentioned above. It was demonstrated that the SVM had a good adaptability and practicability for predicting the productivity of steam flooding production wells. [ABSTRACT FROM AUTHOR]

The productivity of steam flooding production wells is determined by numerous factors. In the paper the dominant influence factors of the productivity have been revealed using Gray Relation Analysis (GRA). Based on the Gray Relation Analysis, a model was established for predicting the productivity of steam flooding production wells by Support Vector Machine (SVM), the training algorithm was used to train the model and predict the productivity of test samples. The results show that high precision can be achieved from SVM, the error committed by the model was about ± 10% which is acceptable in field application. The results of SVM were compared with those of Group Method of Data Handling (GMDH) and Backward Propagation (BP) Artificial Neural Network, it indicated that the prediction precision of SVM was obviously higher than that of the two mentioned above. It was demonstrated that the SVM had a good adaptability and practicability for predicting the productivity of steam flooding production wells. [ABSTRACT FROM AUTHOR]

Background: In recent times there has been a significant change in lifestyle in many parts of the world, with most people experiencing a more sedentary existence combined with an abundance of food. This has resulted in the modern epidemic of obesity and consequent hyperinsulinemia - situations which in women may precipitate expression of fertility problems; effective methods to evaluate the fertility status are required. Ultrasonographic imaging is an effective, easy to use, safe, and readily available noninvasive means to evaluate fertility potential.Objective: Manual recognition of the follicles in terms of area measurement and counting the number of follicles is laborious; often fatigue may lead to error-prone conclusions. The paper attempts an automated classification of the ovaries based on the biomarking done by the physician. Also, biomarked data correlates with the hormones values such as androgen, testosterone and leutinizing hormone.Methods: Despeckled images are segmented by improved active contour with split-Bregman optimization. The features are extracted from images using geometric and intensity method. The significant features selected by particle swarm optimization and dimension reduction by principal component analysis and classification by probabilistic neural network.Results: Proposed probabilistic neural network achieves maximum efficiency of 97% compared to SVM 92% and RBF 88%.Conclusions: The results obtained show that using a very large number of features combined with a feature selection approach allows us to achieve high classification rates. [ABSTRACT FROM AUTHOR]

For solving classification and regression problems, we propose a hybrid system consisting of two phases which work in tandem. In the first phase, particle swarm optimization is employed to train a 3-layered auto associative neural network (henceforth called PSOAANN). In this phase, dimensionality reduction takes place in hidden layer, where the hidden nodes should be less than the input nodes. The outputs from the hidden nodes are then treated as nonlinear principal components (NLPC). They are fed to the second phase where several classifiers and regression methods are invoked. The second phase includes techniques viz., threshold accepting logistic regression (TALR), probabilistic neural network (PNN), group method of data handling (GMDH), support vector machine (SVM) and genetic programming (GP) for classification problems. For regression problems, general regression neural network (GRNN) is used in place of PNN. In addition, support vector machine (SVM), Genetic Programming (GP), GMDH are also employed, as they are versatile. The efficiency of the hybrid is analyzed on five banking datasets namely Spanish banks, Turkish banks, US banks and UK banks and UK credit dataset and five regression datasets viz., Bodyfat, Forestfires, AutoMPG, Boston Housing and Pollution. All the datasets are analyzed using 10 fold cross validation (10 FCV). It turns out that the proposed hybrid yielded higher accuracies across classification and regression problems. [ABSTRACT FROM AUTHOR]

In this paper, a Computer Aided System for microcalcification cluster diagnosis in mammographic images is presented. The method is characterized by three phases. In fact, single microcalcifications are first localized then microcalcifications having a cluster pattern are detected. In the last step the procedure classifies the abnormalities as benign or malignant microcalcification clusters. Features extracted from localized single microcalcifications are fed into a Support Vector Machine classifier to verify the presence of microcalcification cluster, minimizing false positive detections. For the diagnosis purpose, an Artificial Neural Network classifier is implemented which makes use of features extracted from previously detected microcalcification clusters as inputs. The performance of the implemented system is evaluated taking into account the accuracy of both detecting and classifying microcalcification clusters. Adopting the MIAS database as test bench, a sensitivity of about 98.4 at a rate of 0.85 FP/image is achieved in detecting microcalcification clusters. Moreover, the method gets a sensitivity of about 93.5% and an accuracy value equal to 94.2% in classifying the detected microcalcification clusters. The obtained system performance shows its ability of aiding the interpretation of specialists and, consequently, it could be considered as a "second opinion" method. [ABSTRACT FROM AUTHOR]

ChemModLab, written by the ECCR @ NCSU consortium under NIH support, is a toolbox for fitting and assessing quantitative structure-activity relationships (QSARs). Its elements are: a cheminformatic front end used to supply molecular descriptors for use in modeling; a set of methods for fitting models; and methods for validating the resulting model. Compounds may be input as structures from which standard descriptors will be calculated using the freely available cheminformatic front end PowerMV; PowerMV also supports compound visualization. In addition, the user can directly input their own choices of descriptors, so the capability for comparing descriptors is effectively unlimited. The statistical methodologies comprise a comprehensive collection of approaches whose validity and utility have been accepted by experts in the fields. As far as possible, these tools are implemented in open-source software linked into the flexible R platform, giving the user the capability of applying many different QSAR modeling methods in a seamless way. As promising new QSAR methodologies emerge from the statistical and data-mining communities, they will be incorporated in the laboratory. The web site also incorporates links to public-domain data sets that can be used as test cases for proposed new modeling methods. The capabilities of ChemModLab are illustrated using a variety of biological responses, with different modeling methodologies being applied to each. These show clear differences in quality of the fitted QSAR model, and in computational requirements. The laboratory is web-based, and use is free. Researchers with new assay data, a new descriptor set, or a new modeling method may readily build QSAR models and benchmark their results against other findings. Users may also examine the diversity of the molecules identified by a QSAR model. Moreover, users have the choice of placing their data sets in a public area to facilitate communication with other researchers; or can keep them hidden to preserve confidentiality. [ABSTRACT FROM AUTHOR]

This paper presents a hand-shape biometric system based on a novel feature extraction methodology using the morphological pattern spectrum or pecstrum. Identification experiments were carried out using the obtained feature vectors as an input to some recognition systems using neural networks and support vector machine (SVM) techniques, obtaining in average an identification of 98.5%. The verification case was analyzed through an Euclidean distance classifier, obtaining the acceptance rate (FAR) and false rejection rate (FRR) of the system for some K-fold cross validation experiments. In average, an Equal Error Rate of 2.85% was obtained. The invariance to rotation and position properties of the pecstrum allow the system to avoid a fixed hand position using pegs, as is the case in other reported systems. The results indicate that the pattern spectrum represents a good alternative of feature extraction for biometric applications. [ABSTRACT FROM AUTHOR]

The paper is concerned with the application of Support Vector Machine (SVM) to the fault location in the analog electrical circuits. The recognition of fault is based on the measurements of the accessible terminal voltage and current of the network at the set of frequencies. The SVM network is applied as the recognizing system and as the classifier. The important feature of the proposed solution is its high accuracy and great speed of operation. Once the network has been trained, the recognition of fault is achieved immediately, irrespective of the size of the circuit. Thus the solution is suited for real time applications for fault location in electrical circuits. The numerical results of recognition of faulty elements in two different structures of electrical filters are presented and discussed in the paper. [ABSTRACT FROM AUTHOR]

The major drawback of Support Vector Machines (SVMs) consists of the training time, which is at least quadratic to the number of data. Among the multitude of approaches developed to alleviate this limitation, several research works showed that mixtures of experts can drastically reduce the runtime of SVMs. The mixture employs a set of SVMs each of which is trained on a sub-set of the original dataset while the final decision is evaluated throughout a gater. The present work proposes a new support vector mixture in which Sugeno's fuzzy integral is used as a gater to remove the time complexity induced by conventional gaters such as artificial neural networks. Experiments conducted on standard datasets of optical character and face recognition reveal that the proposed approach gives a significant reduction of the runtime while keeping at least the same accuracy as the SVM trained over the whole dataset. [ABSTRACT FROM AUTHOR]

Fingerprint classification reduces the searching time of an automated fingerprint identification system. Since fingerprints have properties of intra-class diversities and inter-class similarities, the ambiguous example causes a difficult problem in the fingerprint classification. In order to address the problem, we have analyzed fingerprints' subclasses with multiple decision templates. It clusters the soft outputs of support vector machines (SVMs) into several sub-classes using the self-organizing maps, and estimates a localized template for each sub-class. For an input fingerprint, the proposed method matches the output vector of SVMs to each template and finally categorizes the sample into the class of the most similar template. Experimental results on the FingerCode dataset demonstrate the effectiveness of the subclass-based approach compared with previous methods. [ABSTRACT FROM AUTHOR]

This paper presents a multi-class Support Vector Machine (SVM) based algorithm for on-line static security assessment of the power systems. The proposed SVM based security assessment algorithm has a very small training time and space in comparison with the traditional machine learning methods such as Artificial Neural Networks (ANN) based algorithms. In addition, the proposed algorithm is faster than existing algorithms. One of the main points, to apply a machine learning method is feature selection. In this paper, a new Decision Tree (DT) based feature selection algorithm has been presented. The proposed SVM algorithm has been applied to New England 39-bus power system. The simulation results show the effectiveness and the stability of the proposed method for on-line static security assessment. The effectiveness of the proposed feature selection algorithm has been investigated, too. The proposed feature selection algorithm has been compared with different feature selection algorithms. The simulation results demonstrate the effectiveness of the proposed feature algorithm. [ABSTRACT FROM AUTHOR]

The selection of a proper auditor is driven by several factors. Here, we use three data mining classification techniques to predict the auditor choice. The methods used are Decision Trees, Neural Networks and Support Vector Machines. The developed models are compared in term of their performances. The wrapper feature selection technique is used for the Decision Tree model. Two models reveal that the level of debt is a factor that influences the auditor choice decision. This study has implications for auditors, investors, company decision makers and researchers. [ABSTRACT FROM AUTHOR]

Microarrays are emerging technologies that allow biologists to better understand the interactions between several pathologic states, at genes level. However, the amount of data generated by these tools becomes problematic when data are supposed to be automatically analyzed (e.g. for diagnostic purposes). In this work, the authors present a novel gene selection method based on Genetic Algorithms and Support Vector Machines (SVMs) for the classification of tissue samples. For such, the authors use an error estimate for SVMs to evaluate each individual's fitness. The proposed method is compared with common used gene selection techniques. Experimental results carried out using public available microarray datasets demonstrated the strength of the approach. [ABSTRACT FROM AUTHOR]

Support Vector Machines constitute a Machine Learning technique originally designed for the solution of 2-class problems. For multiclass applications, several strategies divide the original problem into a set of binary subtasks, whose results are combined. This work introduces the use of Genetic Algorithms to determine binary decompositions of multiclass problems. Experimental results on benchmark and Bioinformatics multiclass datasets indicate the potential of the proposed approach, which is able to produce good multiclass solutions with the use of simple decompositions. [ABSTRACT FROM AUTHOR]

Purpose: In this paper we address the demanding diagnostic problem of classifying tumors according to the degree of their malignancy by investigating the efficiency of Support Vector Machines (SVMs) and Probabilistic neural networks (PNN). Material and methods: 129 cases of urinary bladder carcinomas were diagnosed as high or low-risk according to the WHO grading system. Each case was represented by 36 automatically extracted nuclear features. Two different classification designs based on SVMs and PNNs were tested according to their ability in differentiating superficial urinary bladder carcinomas according to the degree of malignancy. Best feature combination for each classification scheme was obtained performing an exhaustive search in feature space and employing the leave-one-out method. Results: Both classification models (SVM and PNN) resulted in a relatively high overall accuracy of 85.3% and 83.7% respectively. Descriptors of nuclear size and chromatin cluster patterns were participated in both best feature vectors that optimized classification performance of the two classifiers. Conclusion: The good performance and consistency of the SVM and PNN models render these techniques viable alternatives in the diagnostic process of assigning urinary bladder tumors grade. [ABSTRACT FROM AUTHOR]